Inter-office memo
- To
- Everyone choosing an app to run coding agents
- From
- Mo, Chief Executive
- Date
- November 17, 2026, 4:41 AM CST
- Re
- Paseo alternatives
Paseo alternatives in 2026: 9 apps for running coding agents on any machine
Paseo is free, open source and one of the most complete apps for running coding agents from your desk and phone, so leave only if it doesn’t fit. Orca sets up the most agents, free; T3 Code is simpler; Happy puts a teammate in the session; Herdr runs agents in terminals over SSH; Enjoy keeps every lab’s agent going through usage limits without plugins; Claude Code and Paradev run agents in the cloud.
Disclosure: Enjoy is our product. We checked every app below against its maker’s own pages on November 17, 2026, and we say where another app is the better pick, Paseo included.
Paseo alternatives compared
Paseo’s pitch is “Run many coding agents at once, on any machine. From your desk and from your phone.” A daemon on your computer or a server runs the agents, and desktop, phone, web and command-line apps connect to it, through an end-to-end encrypted relay if you turn it on. It’s free and open source, so most people who leave want something it doesn’t do, or does only through a plugin. These are the apps people weigh against it.
| App | Best for | Price, checked November 17, 2026 | Computers | Phone | Agents |
|---|---|---|---|---|---|
| Paseo | Agents on any machine, from a full phone app | Free; hosted Hub free for 50 runs a month, or $15 a seat | Mac, Windows, Linux | iOS, Android, web | Claude Code, Codex, Copilot, OpenCode, Pi, Antigravity, Muse Code, and 30+ through ACP |
| Orca | The most agents set up, free | Free | Mac, Windows, Linux | iOS, Android, in beta | 40 in its docs, and any command-line agent |
| T3 Code | A simpler free app, on a server or your phone | Free | Mac, Windows, Linux | iOS, Android | Claude Code, Codex, Cursor, OpenCode, Grok Build, Antigravity, Kiro, Pi, Muse Code, fx, and ACP agents |
| Happy | Phone-first, with a teammate in the session | Free | Mac, Windows, Linux | iOS, Android, web | Claude, Codex, Grok, Gemini through Antigravity, and ACP agents |
| Herdr | Agents in real terminals, on your SSH machines | Free | Mac (Apple silicon), Windows, Linux | iPhone, iPad, Android | Anything that runs in a terminal |
| Enjoy | Every lab’s agent, through usage limits | A monthly subscription, free for 7 days, with your phone and one computer (pricing) | Mac, Windows, Linux | iPhone app, any browser | Claude Code, Codex, Grok Build, Antigravity, Muse Code, OpenCode |
| Paradev, formerly Superset | Teams on Macs, with cloud workspaces | Free for one user; Pro $20 a user a month; Team $60; Ultra $200 | Mac; Linux experimental | iPhone, iPad, with Pro | 22 built in, and any command-line agent |
| Conductor | A Mac app with cloud sandboxes | Free to use, its docs say; no prices listed | Mac | iOS, experimental | Claude Code, Codex, Cursor, OpenCode |
| Claude Code Desktop | Only Claude, with cloud sessions | Included in Claude Pro ($20 a month), Max ($100) and Ultra ($200) | Mac, Windows; Linux in beta | Claude app | Claude Code |
| The ChatGPT desktop app | Only Codex | On every ChatGPT plan, from Free; Plus is $20 a month | Mac, Windows; Linux in preview | ChatGPT app | Codex |
Prices are per month as each maker lists them, before tax. The independent apps run your agents on your own Claude, ChatGPT or other plans and API keys, which those companies bill for separately. Claude Code Desktop and the ChatGPT app come with those plans.
Why people look for a Paseo alternative
Paseo is free and has a lot built in, so the reasons to move are specific:
- Getting through usage limits takes community plugins. You can add a second account as its own provider entry, which “shows up as a separate option in the app with its own credentials and models,” but you pick it yourself: its docs don’t describe moving work to another account or agent when a plan runs out. Its plugin directory, 559 plugins long, has community ones that do: Usage Auto Resume, to “Continue Codex, Claude, Antigravity, and OpenCode agents automatically after their usage limit resets”; Continuity, which moves work to “a fallback provider after quota or rate-limit failures”; and Paseo Accounts, ZeroSub and Account Switcher, for several subscriptions side by side. Its README warns “Plugins run with access to your daemon machine and inside connected clients; install only code you trust.”
- Your machines do all the work. “Paseo runs a local server called the daemon that manages your coding agents.” Hub adds triggers, but “an agent starts on your daemon,” and Hub “ships no agent of its own.” There’s no managed cloud, so a laptop that sleeps or a server that stops takes its agents with it.
- Its team layer is a small beta. On Hub, “access to the beta stays small and trusted,” and its shared settings and daemons for teams are “Not built yet, listed so you know where this is going.”
Orca
Best for: the most agents set up out of the box, free.
Orca, from Stably, says “Run Claude Code, Codex, and any other coding agent in parallel, each in its own worktree.” Its docs list 40 agents “Orca ships with out of the box,” and “Orca works with any CLI agent.” Its README says “See your GitHub and Linear tasks in one place,” “schedule prompts to run on any repo,” and “Pair a Remote Orca Server, work on any SSH host, or give a worktree its own cloud VM,” with the promise “Your agents keep running when your laptop sleeps.” Its mobile companion for iOS and Android pairs with your desktop to reply to prompts, review source control and switch agent accounts.
Price: free. “Orca is free and open source under the MIT License,” on macOS, including Intel Macs, Windows and Linux.
Watch for: “The mobile companion is in beta,” and “Closing the desktop app drops a direct/LAN session until you reopen it.” Its always-on machines are yours to run: “Orca does not sell managed VPS hosting,” says Ways to run, and the cloud VM option sits “under Settings → Experimental.” And “Orca pre-fills each supported CLI’s permission-bypass flag for new launches,” so agents don’t ask first unless you change it. See Orca alternatives.
T3 Code
Best for: a simpler free app, on your desktop, a server or your phone.
T3 Code calls itself “the open-source control plane for coding agents.” Its README says “It works with Claude Code, Codex, Cursor, OpenCode, Grok Build, Antigravity, Kiro, Pi, Muse Code, fx, and any ACP agent,” “Each thread can work in its own worktree,” and you can “Change models in the middle of a thread.” It has desktop apps for macOS, Windows and Linux, a web app, and “apps for iOS and Android.” Its FAQ says “Your agents run wherever T3 Code runs: your desktop, a server, a VPS.”
Price: “Nothing. There’s no way to pay us to use T3 Code. You bring your own sub.” It’s MIT-licensed, and T3 Connect, which links your phone and machines, “is free too, for now,” for “up to 3 machines” an account.
Watch for: “The phone is a remote. Your computer does the work and needs to stay on.” On T3 Connect’s cost, its FAQ adds “if the bill gets unacceptable we may have to change course.” Its FAQ doesn’t mention schedules, and it skips local models: “Locally runnable models are not capable of meaningful engineering work.” See T3 Code alternatives.
Happy
Best for: following agents from your phone, with a teammate in the same session.
Happy’s original command-line tool “wraps the Claude Code and Codex harnesses directly in your terminal and lets you continue from your phone,” and its README now says “Orchestrate Claude, Codex and Grok in one app.” It runs “Any model: Claude, Codex, Grok, Gemini, and more”: “Gemini runs through Antigravity, and OpenCode or any Agent Client Protocol agent plugs in as a harness.” It has an “End-to-end encrypted mobile app” for iOS, Android and the web, and a desktop app for macOS, Windows and Linux. It’s “Natively multiplayer”: “Invite a colleague or a friend into the session. You both watch the same agent work, and either of you can steer it.”
Price: free. Its homepage says “Free and open source,” under the MIT license, and “Happy adds a harness, not another bill.”
Watch for: “It runs on your own hardware,” so your computer does the work, and its README and homepage don’t mention worktrees or schedules, which Paseo has. Its homepage lists no paid plans.
Herdr
Best for: agents in real terminals, on any machine you reach over SSH.
“Herdr is an app for running coding agents that you shape by asking.” It works with “Anything that runs in a terminal,” and you can “Add any machine you can reach over SSH, like a server, a VM or a cloud sandbox”: “Its agents keep running there while your laptop sleeps or the app is closed.” It runs on “macOS (Apple silicon), Linux (x86_64 and arm64) and Windows (x86_64),” with apps for “iPhone, iPad and Android.”
Price: free. “The app, herdr-core and Herdr TUI are all open source,” under Apache 2.0.
Watch for: its downloads don’t include Intel Macs, and its homepage doesn’t describe worktrees or schedules. “If the machine reboots or herdr-core stops, the programs stop too,” and “Herdr doesn’t restart an agent or resume its work in progress.”
Enjoy
Best for: keeping work going when an agent or a plan runs out, without plugins.
Enjoy runs Claude Code, Codex, Grok Build, Antigravity, Muse Code, and local models through OpenCode, on Mac, Windows and Linux, on the AI plans you already have. Each conversation runs its own agent, and an agent can work in a worktree. Like Paseo, it reaches your computer from your phone: pair your phone with a code, and use the iPhone app or any phone’s browser. What passes between them is end-to-end encrypted, and the relay “can’t read it.” What it adds:
- Keep going through limits, without plugins. When an agent runs out of usage, Enjoy switches to another of your accounts that still has usage, or waits for the limit to reset, then continues. Claude Code, Codex and Grok Build keep several accounts signed in, and each project can use its own.
- Switch agents mid-conversation. When one lab’s agent is the wrong fit or out of usage, another picks up the same conversation: “The new agent gets the conversation so far, including replies from other agents, before it works on your message.”
- A superagent over everything. JOY keeps an eye on all your projects: ask it what needs you, catch up on what your agents did, or start work in any project.
Price: Enjoy is one monthly subscription, and the first seven days are free, with a card to start. It includes your phone and any browser and one computer; each other computer that runs its own projects costs extra each month, and each teammate you invite needs a subscription of their own. See pricing.
Watch for: Enjoy isn’t open source, and it’s a monthly subscription on top of your AI plans once the free week ends, where Paseo is free. It runs six agents, where Paseo adds 30 or more through ACP. It has nothing like Hub: agents don’t start from a GitHub, Slack or Discord mention. It has no cloud of its own, so agents run on a computer where Enjoy is installed, and scheduled workflows run only while that computer is awake with Enjoy open. And it doesn’t stop two conversations from editing the same files: overlapping work needs a worktree, which you ask for.
Paradev, formerly Superset
Best for: teams on Macs who want agents in managed cloud workspaces.
“Superset is now Paradev,” its homepage says, “The open-source workspace for Claude Code, Codex, and any coding agent.” Its supported agents page lists 22, and “Any CLI agent runs in Paradev.” “Cloud workspaces run in a sandbox,” off your computer, with 150 hours per seat each month on Team and 500 on Ultra.
Price: its pricing has a free plan that “covers individuals with 1 user, local workspaces, the desktop app, and the CLI.” Pro, which adds remote access, automations and the phone app, is $20 a user a month, or $16 billed yearly; Team is $60, or $54 yearly; Ultra is $200. It’s under the MIT license.
Watch for: its FAQ says “macOS (Apple Silicon and Intel) is the supported target,” the Linux build “is experimental and untested,” and “Windows is planned with no date.” The phone app “needs iOS 26 or iPadOS 26 or later and Paradev Pro,” and “Android is not available yet.” For local work from your phone, “Keep the computer awake, online, and running Paradev,” and its remote access page doesn’t say whether relay traffic is end-to-end encrypted. See Superset alternatives.
Conductor
Best for: Mac users who want a polished Mac app with cloud sandboxes.
Conductor says “Run parallel Claude Code, Codex, and Cursor agents in isolated workspaces on your Mac,” and its harnesses page adds OpenCode. Its homepage says “Conductor Cloud runs on isolated microVMs with your repos and dependencies pre-installed,” and “Conductor is now multiplayer.”
Price: “Conductor is free to use. We do not bill or resell model usage.” It lists no prices.
Watch for: “Conductor is not available for Windows or Linux yet,” says its installation page. Its iOS remote control “is experimental,” and “Conductor must be open on your Mac, and your Mac must be awake and online.” It’s joining Meta, which says “Plans have changed! We’ll still be developing Conductor,” and “We’ll have something new for you to use early next year.” See Conductor alternatives.
Claude Code Desktop
Best for: people who only use Claude, with sessions that keep going with your computer off.
Claude Code is the Code tab of Anthropic’s Claude desktop app, where you can open new sessions “to work on multiple tasks in parallel.” Its docs say “Cloud sessions run on Anthropic-managed infrastructure by default and continue even if you close the app or shut down your computer,” and you can follow them “from claude.ai/code or the Claude mobile app.” With Remote Control, a session on your computer appears “in the Claude apps on devices signed in to your claude.ai account.” It runs on Mac and Windows, with Linux in beta.
Price: Claude Code is on Anthropic’s paid plans, per its pricing: Pro at $20 a month, or $17 billed yearly, Max at $100 and Ultra at $200. The Free plan doesn’t include it.
Watch for: it runs only Claude, and “By default, Worktree is off and Claude works in the project folder itself,” so local sessions share one folder unless you turn it on.
The ChatGPT desktop app, for Codex
Best for: people who only use OpenAI.
OpenAI’s desktop app is where Codex runs on your computer: “Run projects in parallel, work with files, use your computer, and keep long-running work moving from one desktop workspace.” You can “Download ChatGPT for macOS, Windows, or Linux,” and on Linux it’s “available in preview.” With remote connections, the ChatGPT app on an iPhone or Android phone works with Codex chats on a connected Mac or Windows computer.
Price: on every ChatGPT plan, from Free. Go is $8 a month, Plus $20, and Pro starts at $100. Plus adds “Codex on the web, in the CLI, in the IDE extension, and on iOS.”
Watch for: it runs only Codex, and Free and Go get “GPT-6 Luna at Standard speed in the desktop app, subject to rollout.” For the phone, the computer must be “awake, online, and signed in to the same account and workspace,” and if it sleeps, “remote access stops until it’s available again.”
When to stay with Paseo
For most people, stay. Paseo is free under Apache 2.0, “with no telemetry, tracking, or required Paseo account,” and no other app here documents all of this at once: “The native mobile app has full feature parity with desktop,” its relay is end-to-end encrypted (“Paseo can’t read your traffic”), voice “runs locally on your device by default,” and its schedules start “a new agent for you on a cron cadence.” Its orchestration lets agents “split work across providers and machines.”
Stay too if you’re comfortable choosing plugins. The limits and accounts that send some people elsewhere are covered by community plugins in its directory; read each one’s code first, as Paseo’s README advises. Leave if you’d rather have that built in, need agents that run with every machine of yours off, or want a team product that’s out of beta.
How we compared
We read each maker’s homepage, pricing, download and docs pages, and their GitHub repositories, on November 17, 2026, and every one is linked below. We quote their own words for the limits, and we don’t count features a maker hasn’t documented. Paseo’s homepage names OMP among its agents and its README names Copilot; we list the README’s, plus the more than 30 ACP agents its supported providers page catalogs. Orca’s homepage names 27 agents; we use its docs, which list 40 set up out of the box. Roundups at OpenAlternative and AgentsRoom list more apps, including Orchestrator.inc, Emdash, Synara, Jean and AgentsRoom itself, which we didn’t compare here. This category changes weekly, so we recheck this guide against its sources at least once a month, and the date at the foot of the page says when we last did.
14.Questions.
What’s the best alternative to Paseo?
- It depends on why you’re leaving. Orca sets up 40 agents and is free; T3 Code is a simpler free app; Happy is phone-first and lets a teammate into the session; Herdr runs agents in real terminals on machines you reach over SSH; Enjoy keeps every lab’s agent going through usage limits without plugins; and Paradev, Conductor or Claude Code run agents in the cloud with your computer off.
Is Paseo free?
- Yes. “Paseo is free and open source,” under the Apache 2.0 license, with no Paseo account required. Only Paseo Hub, which starts agents from GitHub, Slack and Discord, has a paid plan, $15 a seat a month; it’s free hosted for 50 agent runs a month and one seat, and free to host yourself.
Can Paseo switch accounts when Claude or Codex hits its limit?
- Only through community plugins. You can add accounts as separate provider entries and pick one yourself; for switching or resuming on its own, its plugin directory lists plugins such as Usage Auto Resume, Paseo Accounts and Account Switcher, and Paseo’s README warns that plugins run with access to your daemon machine, so “install only code you trust.” Enjoy switches to another of your accounts, or waits for the reset, on its own.
Which Paseo alternatives keep agents running with my computer off?
- Claude Code’s cloud sessions “continue even if you close the app or shut down your computer,” and Paradev’s cloud workspaces, from its $60 Team plan, and Conductor Cloud run agents for you too. Orca, Herdr, T3 Code and Paseo itself can run agents on a server or another machine, but one you keep running.
Do I still need a Claude or ChatGPT plan?
- Yes. Paseo and the independent apps here run agents on your own Claude, ChatGPT or other plans and API keys, which those companies bill for separately. Claude Code Desktop and the ChatGPT desktop app come with those plans.